Concentration of power is Amodei's name for the structural feature of the AI industry that he identified in 'The Adolescence of Technology' as historically unprecedented. The companies building the most powerful AI systems are accumulating a form of power that has no precedent: the power to build systems that could replace or augment human cognitive labor across the entire economy. This power is concentrated in a small number of organizations, led by a small number of individuals, operating in a regulatory vacuum. Amodei predicted that AI could create personal fortunes well into the trillions for a powerful few. The concentration is not the result of anyone's malice but the predictable consequence of incentive structures that reward capability investment and fail to reward safety investment, compounded by network effects and scaling economics.
The mechanisms driving concentration are structural. Developing frontier AI requires capital on a scale that few organizations can assemble — training a frontier model costs hundreds of millions of dollars, with the next generation likely to cost billions. This capital requirement creates a natural oligopoly: only organizations with access to extraordinary funding can participate at the frontier. The result is that the decisions about what capabilities to develop, how to deploy them, and what values to embed are being made by a handful of organizations whose leaders were not elected and whose authority derives entirely from their commercial success.
Network effects compound the concentration. Users who adopt a particular AI system accumulate data, develop workflows, and build dependencies that make switching costly. Developers who build on a particular platform create applications that lock users in further. The result is winner-take-all or winner-take-most dynamics in which early leaders capture increasing shares of the market and the resources required to stay at the frontier. The economic logic is well understood — network effects are a standard feature of platform economics — but its application to a technology as consequential as AI produces concentration at civilizational scale.
Amodei's response to this concentration has been publicly to advocate constraints on his own industry. In November 2025, on 60 Minutes, he described his deep discomfort with AI decisions being made by a few companies and few people. The concentration of power had happened almost overnight and almost by accident. He argued that AI should be more heavily regulated, with fewer decisions left to technology company heads. The CEO of a company that would be constrained by the regulation he was advocating publicly called for that constraint because he believed it necessary for responsible development.
The wealth pledge accompanying 'The Adolescence of Technology' — the commitment by Amodei and six other Anthropic co-founders to donate eighty percent of their wealth — was explicitly framed as a response to the concentration risk. Amodei was clear that the pledge was not a solution to the structural problem; no individual act of philanthropy could address incentive systems at industry scale. The pledge was a signal, an argument that the race for AI supremacy did not have to be a race for personal enrichment. Whether the signal would be heard above the noise of the race was unclear, but the signal was sent.
The concept is developed most fully in Amodei's January 2026 essay 'The Adolescence of Technology,' though concerns about concentration of power had appeared in his public statements since Anthropic's founding. The specific framing as 'historically unprecedented' draws on comparative analysis of previous technology transitions — the railroad, the automobile, electricity, the internet — each of which produced concentrated industries but none at the scale or speed of AI.
The concept intersects with broader analyses of concentrated economic power, including work by Branko Milanovic on the plutocratic bias of contemporary capitalism and analyses of the broligarchs by Al Gore and others.
Historically unprecedented. The scale and speed of AI industry concentration exceed previous technology transitions — railroad, automobile, electricity, internet.
Structural, not malicious. Concentration results from capital requirements, network effects, and incentive structures, not from bad actors.
Trillion-dollar fortunes. AI could create personal wealth on a scale previously unimaginable, concentrated in a small number of individuals.
Regulatory vacuum. The concentration is occurring in the absence of governance frameworks adequate to the stakes.
CEO advocacy of constraints. Amodei's public advocacy of regulation that would constrain his own company represents a rare form of institutional self-awareness.
Critics argue that Anthropic's own structure — a for-profit company accepting billions in investment — contributes to the concentration Amodei decries, and that his advocacy of regulation is therefore either incoherent or self-serving (since regulation often entrenches incumbents). Defenders argue that the alternative of ceding frontier development to less safety-focused companies would be worse, and that Amodei's willingness to publicly advocate constraints on his industry represents the most that can be expected from a leader operating within the existing economic structure.